Fix for Llama4 Maverick performance drop#904
Merged
wpyszka merged 3 commits intoFeb 2, 2026
Merged
Conversation
Signed-off-by: Jan Kaniecki <jan.kaniecki@intel.com>
jkaniecki
requested review from
mgawarkiewicz-intel,
piotrbocian and
wpyszka
as code owners
Contributor
There was a problem hiding this comment.
Pull request overview
This PR reverts a performance-degrading patch for Llama4 Maverick models by removing the patch_llama4_get_attn_scale function and its invocation. The patch was causing recompilations that significantly reduced performance.
Changes:
- Removed the
patch_llama4_get_attn_scalefunction that was modifying attention scale behavior for Llama4 models - Removed the call to
patch_llama4_get_attn_scalefromapply_model_specific_patches
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
mgawarkiewicz-intel
approved these changes
Feb 2, 2026
✅ CI PassedAll checks passed successfully against the following vllm commit: |
Collaborator
|
Is this needed on main? |
Luca-Calabria
added a commit
to Luca-Calabria/vllm-gaudi
that referenced
this pull request
Feb 6, 2026
Signed-off-by: Luca Calabria <luca.calabria@intel.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This is a fix for Maverick performance drop - t.compile does not handle functions with methods as inputs, so to avoid recompilations we need to declare a scale function directly.